sign in sign up
<<
Home , Resistin

European Journal of Endocrinology (2003) 149 231–238 ISSN 0804-4643

EXPERIMENTAL STUDY Differential expression of and resistin in 3T3-L1 treated with -a Yin Li, Kazuhito Totsune1, Kazuhisa Takeda, Kazumichi Furuyama, Shigeki Shibahara and Kazuhiro Takahashi Department of Molecular Biology and Applied Physiology and 1Department of , Tohoku University School of Medicine, 2-1 Seiryo-machi, Aoba-ku, Sendai, Miyagi 980-8575, Japan (Correspondence should be addressed to K Takahashi; Email: [email protected])

Abstract Design: It has recently been shown that deficiency of adrenomedullin (AM), a potent vasodilator pep- tide, leads to resistance. We studied expression of AM in NIH 3T3-L1 adipocytes and compared it with expression of resistin, an -derived that is proposed to cause . Moreover, we studied the effects of tumor necrosis factor-a (TNF-a), a known mediator of insulin resistance, on the expression of AM and resistin in 3T3-L1 adipocytes. Methods: 3T3-L1 cells were induced to differentiate to adipocytes by insulin, dexamethasone and 3-isobutyl-1-methylxanthine. Expression of AM mRNA and resistin mRNA was examined by North- ern blot analysis. Immunoreactive AM in the medium was measured by RIA. Results: AM mRNA was expressed in preadipocytes, but barely detectable in adipocytes. Immuno- reactive AM was detected in the medium of both preadipocytes and adipocytes, with about 2.5 times higher levels found in preadipocytes. In contrast, resistin mRNA was expressed in adipocytes, whereas it was not detected in preadipocytes. Treatment with TNF-a increased AM expression in both adipocytes and preadipocytes, whereas it decreased resistin mRNA levels in adipocytes. Conclusions: The present study has shown that AM expression was down-regulated and resistin expression was up-regulated during adipocyte differentiation of 3T3-L1 cells. TNF-a acted as a potent negative regulator of resistin expression and a potent positive regulator of AM expression in adipocytes, raising the possibility that in addition to its known actions in causing insulin resistance, TNF-a may also have actions against insulin resistance through AM and resistin.

European Journal of Endocrinology 149 231–238

Introduction (15). There have been no reports, however, on the expression of AM in adipocytes. has recently been considered to play not Resistin is a recently discovered only a role in lipid , but also active endo- specifically secreted by adipose tissue, and has been crine roles in secreting a variety of bioactive molecules, suggested to be involved in the pathogenesis of insulin such as (1), (2, 3) and tumor necro- resistance, which may link to diabetes (16). sis factor-a (TNF-a) (4). Adrenomedullin (AM) is a Resistin is a 94- polypeptide containing 11 potent vasodilator peptide consisting of 52 amino Cys residues, and secreted as a disulfide-linked dimer, acids, which was originally isolated from human with a single Cys residue being required for dimerization pheochromocytoma (5). Shimosawa et al. have recently (17). Insulin-sensitizing thiazolidinediones decrease reported that insulin resistance had developed in resistin expression through the pathway of peroxi- 55-week-old heterozygous AM knockout mice (6), some proliferator-activated receptor-g (PPAR-g) (16, 18, indicating that AM deficiency leads to insulin resistance 19). The administration of resistin impaired glucose tol- in aged mice. Plasma levels of AM were elevated in erance and insulin action in mice, whereas neutraliz- patients with diabetes mellitus (7, 8). AM is expressed ation of resistin by anti-resistin antibody improved in the pancreatic islets and inhibits insulin secretion hyperglycemia and insulin resistance in mice fed a in a dose-dependent manner (9, 10). AM appears to high- diet. Resistin antagonized insulin-stimulated be ubiquitously expressed in many types of cells (11), glucose transport in cultured NIH 3T3-L1 adipocytes such as vascular endothelial cells (12), vascular and pretreatment of 3T3-L1 adipocytes with anti-resis- smooth muscle cells (13), fibroblasts (14) and neurons tin antibody augmented insulin-stimulated glucose

q 2003 Society of the European Journal of Endocrinology Online version via http://www.eje.org

Downloaded from Bioscientifica.com at 09/29/2021 03:55:47PM via free access 232 Y Li and others EUROPEAN JOURNAL OF ENDOCRINOLOGY (2003) 149 transport (16). On the other hand, some recent reports resistin and AM in adipocytes, 3T3-L1 adipocytes were on human studies have cast doubt on the relevance of treated with recombinant murine resistin (10210, the rodent data on resistin to human obesity and studies 1029 and 1028 mol/l) or mouse AM (1029,1028 and of resistin have been inconsistent 1027 mol/l). Expression of AM mRNA and resistin (20–22). It has been shown that insulin is a positive mRNA was examined by Northern blot analysis. IR-AM regulator of resistin gene expression (23, 24), whereas in the medium was measured by RIA. a b-adrenergic receptor or TNF-a suppresses resistin expression (25–27). RNA extraction and Northern blot analysis In the present study, we studied expression of AM during adipocyte differentiation of 3T3-L1 cells, and Total cellular RNA was isolated from 3T3-L1 cells using compared it with expression of resistin. Moreover, we TRIZOL reagent (Life Technologies, Gaithersburg, MD, studied the effects of TNF-a as well as two other cyto- USA) according to the manufacturer’s instructions. kines, interleukin-1b (IL-1b) and interferon-g (IFN-g), For Northern blot analysis of resistin, AM and glyceral- on expression of AM and resistin in 3T3-L1 adipocytes. dehyde-3-phosphate dehydrogenase (G3PDH) (as an TNF-a is secreted not only by activated macrophages internal control), 15 mg total RNA were electrophor- and lymphocytes, but also by adipocytes, and has esed on 1% agarose gels containing 16.6% formal- been implicated as one of the factors responsible for dehyde. RNA was then transferred to nylon insulin resistance (4). membranes (Zeta-Probe blotting membranes; Bio-Rad Lab, Hercules, CA, USA) and covalently bonded to the membrane by exposure to UV light. Materials and methods Northern probes for AM and resistin were prepared by Materials RT-PCR. Total RNA (6 mg) prepared from adrenal (for AM) or 3T3-L1 adipocytes (for resistin) was reverse tran- NIH 3T3-L1 cells (28, 29) were obtained from the scribed as previously reported (30). One microliter of the Health Science Research Resource Bank, Osaka, reaction mixture was subjected to PCR. The sense primer Japan. Human TNF-a, human IL-1b and recombinant for rat AM was 50-TTCGAGTCAAACGCTACCGC-30 murine resistin were obtained from PeproTech EC Ltd (nucleotides 266–285) and the anti-sense primer was (London, UK). Recombinant human IFN-g (IFN-g-1a, 50-ATCAGGCGCTCTCCACCTTA-30 (complementary to Imunomax-g) was a gift from Shionogi Co. (Osaka, nucleotides 1102–1121) (31). The sense primer for Japan). Human g-insulin, dexamethasone and 3-isobu- resistin was 50-TTGTGTCCTGCTAAGTCCTC-30 (nucleo- tyl-1-methylxanthine were from Wako Chemical Co. tides 47–66) and the antisense primer was 50-ACATC- (Osaka, Japan). Mouse AM was obtained from Phoenix AGGAAGCGACCTGCA-30 (complementary to nucleo- Pharmaceuticals, Inc. (Belmont, CA, USA). Other pep- tides 412–431) (Gene Bank accession No. tides including human AM were obtained from Peptide AF323080). The PCR was performed as follows: AM, Institute Co. (Osaka, Japan). 40 cycles at 94 8C for 1 min, 60 8C for 1.5 min, 72 8C for 2 min; and resistin, 35 cycles at 96 8C for 30 s, Cell culture and differentiation 45 8C for 1 min, 72 8C for 2 min. The amplified cDNA fragments were inserted into the EcoRV site of the 3T3-L1 cells were cultivated in Dulbecco’s modified pBluescript I KS vector (Stratagene, La Jolla, CA, Eagle’s medium (DMEM) containing 10% (v/v) fetal USA), yielding the subclones pBS-rAM10 and pBS- bovine (FBS). Confluent cells were induced to mRes respectively. The nucleotide sequence was deter- differentiate to adipocytes by culturing for 48 h in mined using an autosequencer (Perkin-Elmer, Chiba, DMEM containing 10% FBS, 1 mg/ml insulin, Japan) and was confirmed to be identical to the regis- 1 mmol/l dexamethasone and 0.5 mmol/l 3-isobutyl-1- tered sequence of rat AM (Gene Bank accession methylxanthine. The cells were then cultured in No. AF323080) (31) and mouse resistin. The DMEM containing 10% FBS for another 24 h. The HindIII/NaeI fragment of pBS-rAM10 (vector/443) medium was replaced by DMEM with 10% FBS, con- and the HindIII/EcoRI fragment of pBS-mRes (vector/ taining either TNF-a, IL-1b or IFN-g. All test reagents vector) were labeled with [a-32P]dCTP with a BcaBest were diluted into culture medium. The 3T3-L1 cells Labeling Kit (Takara Biomedicals, Shiga, Japan). The cultured in parallel in DMEM with 10% FBS, but with- probe for G3PDH was the NcoI/PstI fragment derived out insulin, dexamethasone and 3-isobutyl-1-methyl- from rat 720 bp G3PDH cDNA fragment (nucleotides xanthine, were used as preadipocytes. The cells were 5–104 and 368–987) subcloned pGEM-T vector harvested for RNA extraction and the culture media (Promega, Madison, WI, USA) (32). were collected for the measurement of immunoreactive Blots were prehybridized for 4 h at 428C in a solution AM (IR-AM). The experiments were performed in five containing 50% formamide, 5 £ standard saline citrate dishes per each treatment. (SSC), 5 £ Denhardt’s solution, 200 mg/ml heat- To examine the effects of exogenously added denatured salmon sperm DNA and 1% SDS. Hybridiz- recombinant resistin or synthetic AM on expression of ation was carried out for at least 16 h at 428Cinan

www.eje.org

Downloaded from Bioscientifica.com at 09/29/2021 03:55:47PM via free access EUROPEAN JOURNAL OF ENDOCRINOLOGY (2003) 149 Adrenomedullin and resistin in adipocytes 233 identical solution containing 1–3 £ 106 d.p.m./ml of ANOVA followed by Fisher’s protected least significant labeled probes. The blots were washed twice with difference test. 2 £ SSC and 0.1% SDS at 65 8C for 5 min and twice with 0.1 £ SSC and 0.1% SDS at 65 8C for 30 min. Radioactive signals were detected by exposing the Results filters to X-ray film (X-AR5; Eastman Kodak, Rochester, NY, USA) or with a Bioimage Analyzer (BAS1500; Expression of resistin and AM mRNAs in Fujifilm, Tokyo, Japan). The radioactivity correspond- 3T3-L1 cells, and effects of TNF-a, IFN-g ing to each band was determined using a Bioimage and IL-1b Analyzer. Microscopical examinations confirmed that 3T3-L1 cells showed morphological properties of adipocytes, Peptide extraction and RIA such as accumulation of fat droplets, after the treat- ment with insulin, dexamethasone and 3-isobutyl-1- in culture media were extracted using Sep-Pak methylxanthine. Such changes were not observed in C18 cartridges (Waters, Milford, MA, USA) as pre- control preadipocytes. viously reported (30). The extract was reconstituted AM mRNA was expressed in 3T3-L1 preadipocytes in assay buffer (0.1 mol/l phosphate buffer (pH 7.5) (control of preadipocytes in Fig. 1), whereas it was containing 0.1% (w/v) BSA, 0.2% (v/v) Triton X-100 very weak or undetected in adipocytes (control of adi- and 0.1(w/v) sodium azide) and assayed. pocytes in Fig. 1). 3T3-L1 adipocytes and preadipo- Mouse AM was used as a standard. 125I-human AM cytes were treated with TNF-a (10 ng/ml), IFN-g prepared by the chloramine T method was used as a (100 U/ml) or IL-1b (10 ng/ml) for 24 h. Treatment radioligand (33). The antiserum (No. 103) (15) against with TNF-a greatly augmented AM mRNA expression human AM, which cross-reacted with mouse AM, was levels in both preadipocytes and adipocytes (Fig. 1). In used at a final dilution of 1:20 000. The sample or the preadipocytes, IL-1b slightly augmented AM mRNA standard peptide (200 ml) was incubated with the anti- expression, whereas IFN-g did not affect AM mRNA serum (200 ml) at 4 8C for 48 h and then 100 ml 125I- expression. AM (approximately 4000 c.p.m./0.1 ml) were added to On the other hand, resistin mRNA was detected in each sample. After a further 48 h incubation at 4 8C, differentiated adipocytes but not in preadipocytes by 100 ml 5% anti-rabbit IgG raised in goat were added. Northern blot analysis. TNF-a treatment inhibited After 1 h incubation, the samples were centrifuged at resistin mRNA level by about 60% compared with 2300 g for at least 1 h and the supernatant was aspi- the untreated adipocyte control. In contrast, IFN-g or rated. The pellets were counted in a g-counter. Intraas- say and interassay coefficients of variation were 5.6 and 5.8% respectively. The assay could detect changes of 10^3.4 fmol/tube (mean^S.D., n ¼ 5) from zero with 95% confidence. There was no significant cross-reac- tion (,0.01%) with other peptides including gene-related peptide (CGRP), Y, somato- , -1, atrial (CGRP), corticotropin-releasing hormone and vasoactive intesti- nal polypeptide. Chromatographic characterization of IR-AM in the conditioned medium was performed by reverse-phase HPLC using mBondapak C18 column (3.9 £ 300 mm: Waters), as previously reported (30, 33). The con- ditioned medium was extracted with a Sep-Pak C18 cartridge, reconstituted in 0.1% (v/v) trifluoroacetic acid and loaded onto the column. HPLC was performed with a linear gradient of acetonitrile containing 0.1% (v/v) trifluoroacetic acid from 10 to 60% at a flow rate of 1 ml/min per fraction over 50 min. Each fraction (1 ml) was collected, air-dried and assayed. Figure 1 Northern blot analysis of AM and resistin mRNAs in 3T3-L1 preadipocytes and adipocytes treated with IFN-g Statistics (100 U/ml), TNF-a (10 ng/ml) or IL-1b (10 ng/ml) for 24 h. Each lane contains 15 mg total RNA. Bottom panel shows the Data are expressed as means^S.E.M., unless otherwise expression of G3PDH mRNA as an internal control. Data are from stated. Statistical analysis was performed by one-way one of three independent experiments with similar results.

www.eje.org

Downloaded from Bioscientifica.com at 09/29/2021 03:55:47PM via free access 234 Y Li and others EUROPEAN JOURNAL OF ENDOCRINOLOGY (2003) 149

IL-1b did not significantly influence resistin mRNA in the position of mouse AM was larger in 3T3-L1 expression in 3T3-L1 adipocytes (Fig. 1). adipocytes treated with TNF-a (Fig. 3B).

Effects of TNF-a, IFN-g and IL-1b on IR-AM Time-course effects of TNF-a on the expression levels in the medium of 3T3-L1 cells of AM and resistin IR-AM was detected in the media of both preadipo- AM mRNA expression was induced by 10 ng/ml TNF-a cytes and adipocytes, with higher levels found in as early as 3 h, and remained to be increased up to preadipocytes (preadipocytes, 90.48^8.80 fmol/105 24 h in 3T3-L1 adipocytes (Fig. 4A). Fig. 4B shows cells per 24 h, vs adipocytes, 36.21^1.88 fmol/105 time-course effects of TNF-a on IR-AM concentrations cells per 24 h ¼ 172.72^8.97 pmol/l) (Fig. 2). On the other hand, IR-AM was not detected in uncondi- tioned medium (,25 pmol/l), indicating the active secretion of IR-AM from 3T3-L1 preadipocytes and adipocytes. IR-AM levels in the media were significantly elevated by 24 h treatment with 10 ng/ml TNF-a of both preadi- pocytes and adipocytes (Fig. 2), and this elevation was more obvious in preadipocytes than in adipocytes. IL- 1b increased IR-AM levels in the media of only preadi- pocytes. IFN-g had no significant effects on IR-AM levels in the media of either 3T3-L1 preadipocytes or adipocytes. IR-AM in the medium of adipocytes with or without TNF-a treatment was characterized by reverse phase HPLC. As shown in Fig. 3, two major peaks eluting in the positions of mouse AM and mouse AM with oxidized were observed. In addition to these two peaks, several minor peaks were observed and their identity needs further investigations. A peak eluting

Figure 2 IR-AM concentrations in the medium of 3T3-L1 preadi- pocytes and adipocytes treated with IFN-g (100 U/ml), TNF-a Figure 3 Reverse phase HPLC of IR-AM in the medium of (10 ng/ml), or IL-1b (10 ng/ml) for 24 h. Data are means^S.E.M. 3T3-L1 adipocytes. (A) Without TNF-a treatment and (B) with (n ¼ 5). Statistical analysis was performed by one-way ANOVA TNF-a treatment (10 ng/ml) for 24 h. The arrows indicate the followed by Fisher’s protected least significant difference test. elution position of mouse AM (mAM) and mouse AM with oxidized **P , 0.01, ***P , 0.0001 (compared with untreated control within methionine (oxi mAM). Dotted lines show a gradient of the same group). acetonitrile (%).

www.eje.org

Downloaded from Bioscientifica.com at 09/29/2021 03:55:47PM via free access EUROPEAN JOURNAL OF ENDOCRINOLOGY (2003) 149 Adrenomedullin and resistin in adipocytes 235

On the other hand, TNF-a suppressed resistin mRNA expression at a concentration as low as 1 ng/ml and the maximal inhibition was found at a concentration of 10 ng/ml.

Effects of exogenously added synthetic mouse AM or recombinant murine resistin on expression of AM and resistin in 3T3-L1 adipocytes Treatment of 3T3-L1 adipocytes with mouse AM (1029,1028 and 1027 mol/l) had negligible effects on the expression of resistin mRNA and AM mRNA (data not shown). Treatment of 3T3-L1 adipocytes with recombinant murine resistin (10210,1029 and 1028 mol/l) did not cause noticeable changes in the expression of resistin mRNA nor AM mRNA (data not shown). It had also no significant effects on IR-AM con- centration in the medium of 3T3-L1 adipocytes. Thus, resistin and AM appear to have no significant effects on their own expression or on that of each other.

Figure 4 Time-course effects of TNF-a (10 ng/ml) on AM Discussion and resistin expression in 3T3-L1 adipocytes. (A) Northern blot analysis. 3T3-L1 adipocytes were cultured with (þ) or without Insulin resistance is frequently associated with obesity. (2) TNF-a (10 ng/ml). Each lane contains 15 mg total RNA. However, the precise mechanisms linking obesity to Bottom panel shows the expression of G3PDH mRNA as an insulin resistance are still unclear. Recent studies internal control. Data are from one of three independent suggest that various adipocyte-derived factors regulate experiments with similar results. (B) RIA of AM. 3T3-L1 adipocytes were cultured with (þ) or without (2) TNF-a insulin sensitivity. Resistin has been identified as a (10 ng/ml) for 6–24 h and IR-AM concentrations in the media new adipocyte-derived peptide hormone induced were measured by RIA after extraction. Data are means^S.E.M. during , and implicated in causing insulin (n ¼ 5). Statistical analysis was performed by one-way ANOVA resistance (16, 23), although some recent reports have followed by Fisher’s protected least significant difference test. cast doubt on the relationship of resistin to insulin *P , 0.05, **P , 0.01, ***P , 0.0001. resistance (20–22). It has been shown that AM deficiency leads to insulin resistance using aged hetero- zygous AM knockout mice (6). Nishimatsu et al. in the culture media of 3T3-L1 adipocytes. IR-AM con- reported that AM induced activation of the phospha- centrations in the media were significantly higher in tidylinositol 3-kinase/Akt pathway in the rat aortic TNF-a-treated 3T3-L1 adipocytes than in untreated endothelium (34). It is therefore plausible that AM 3T3-L1 adipocytes at all the three time points. acts on muscle or as a positive regulator in the TNF-a suppressed resistin mRNA expression in a insulin signaling pathway, such as the phosphatidyl- time-dependent manner. Significant inhibition was inositol 3-kinase/Akt pathway. The present study has detectable from 6 h after 10 ng/ml TNF-a treatment shown that adipocyte differentiation is accompanied and reached a maximum 24 h after the treatment by down-regulation of AM expression and up-regu- (Fig. 4A). lation of resistin expression, raising the possibility that low expression of AM and high expression of resis- Dose–response effects of TNF-a on the tin in adipocytes may be related to the higher occur- expression of AM and resistin rence of insulin resistance in obese subjects. TNF-a is secreted by adipocytes, and antagonizes the Figure 5A shows the dose–response effects of TNF-a stimulation of glucose uptake by insulin. TNF-a inhib- on the expression of AM and resistin mRNAs. TNF-a ited insulin-mediated tyrosine phosphorylation of the increased AM mRNA expression dose-dependently in and insulin receptor substrate 1 3T3-L1 adipocytes (Fig. 5A). IR-AM levels in the (35–37). TNF-a down-regulates glucose transporter medium were increased by the 24 h treatment with GLUT4 gene expression in cultured adipocytes and TNF-a dose-dependently (Fig. 5B). A significant myocytes (38–41). Thus, TNF-a is supposed to be increase in IR-AM was observed in the adipocytes trea- one of the major causes of insulin resistance. We ted with 1–10 ng/ml TNF-a. hypothesized that TNF-a is related to insulin resistance

www.eje.org

Downloaded from Bioscientifica.com at 09/29/2021 03:55:47PM via free access 236 Y Li and others EUROPEAN JOURNAL OF ENDOCRINOLOGY (2003) 149

cultured human astrocytes (42), whereas it decreased the AM expression in T98G human glioblastoma cells (43), and had no noticeable effects in F0202 human retinal pigment epithelial cells (44). The effects of TNF-a on the AM expression in 3T3-L1 cells were simi- lar to those in vascular smooth muscle cells (13) and human astrocytes (42). It has been proposed that TNF-a expression is increased in adipose tissue of obese subjects. It is there- fore possible that AM expression in adipose tissue of obese subjects is rather enhanced by TNF-a and other humoral factors. Furthermore, plasma concentrations of IR-AM were elevated in patients with diabetes melli- tus (7, 8). Since obesity is common in patients with dia- betes mellitus, our findings have raised the possibility that AM secreted by adipose tissue contributed to the elevated plasma IR-AM concentrations in diabetic patients. Further studies are therefore required to clar- ify whether or not AM expression in adipose tissue is enhanced in the obese state. AM is a potent vasodilator peptide (5). AM secreted by adipocytes may act on vascular vessels and regulate a local circulation in adipose tissue. Receptor activity- modifying protein-1, -2 and -3, which form the AM receptor or the CGRP receptor together with calcitonin receptor-like receptor (45), are expressed in adipose Figure 5 Dose–response effects of TNF-a on AM and resistin tissue (46). Another possible target of AM secreted by expression in 3T3-L1 adipocytes. (A) Northern blot analysis. 3T3- adipocytes may therefore be adipocytes themselves. L1 adipocytes were treated with TNF-a for 24 h at indicated con- centrations. Each lane contains 15 mg total RNA. Bottom panel Biological effects of AM on adipocytes, however, shows the expression of G3PDH mRNA as an internal control. remain to be determined. Data are from one of three independent experiments with similar IFN-g has been shown to affect fat metabolism and results. (B) RIA of AM. Data are means^S.E.M. (n ¼ 5). Statistical adipocyte gene expression (47–50). It has been demon- analysis was performed by one-way ANOVA followed by Fisher’s strated that IFN-g treatment results in the development protected least significant difference test. **P , 0.01, ***P , 0.0001 (compared with untreated controls). of insulin resistance through down-regulation of PPAR-g (51). IL-1b stimulates lipolysis and inhibits lipogenesis (52). In the present study, however, IFN-g also through its paracrine or autocrine actions on adi- or IL-1b did not affect AM or resistin gene expression pocytes, in addition to actions on other tissues such as in 3T3-L1 adipocytes. muscle. Contrary to our expectation, TNF-a increased In the present study, we have shown that AM expression of AM but suppressed expression of resistin expression was down-regulated and resistin expression in 3T3-L1 adipocytes. The suppression of resistin was up-regulated during adipocyte differentiation of mRNA expression by TNF-a is consistent with previous 3T3-L1 cells. We also demonstrated roles for TNF-a reports (26, 27). Thus, these paracrine or autocrine as a potent negative regulator of resistin expression actions of TNF-a in adipocytes appear to be against and a potent positive regulator of AM expression in adi- insulin resistance. pocytes. Studies on expression of AM and resistin in The present study has shown for the first time the adipose tissue of lean and obese subjects or animal production and the secretion of AM by adipocytes, models may further clarify their significance in the although its amount was smaller than that by preadipo- pathogenesis of insulin resistance and other obesity- cytes. AM mRNA expression was very low or undetected related . in untreated 3T3-L1 adipocytes, whereas IR-AM was detected in the culture medium at significantly higher levels than in unconditioned medium. This may be Acknowledgements due to a higher sensitivity of the RIA. HPLC analysis confirmed that IR-AM in the culture media mainly con- This work was supported in part by a Grant-in-aid sisted of AM and AM with oxidized methionine, which for Scientific Research (B) (No. 13470030) and may be generated during the extraction procedure. a Grant-in-aid for Scientific Research on Priority It was reported that TNF-a stimulated AM expression Areas (A) (No. 13035005) from the Ministry of in cultured vascular smooth muscle cells (13), and Education, Science, Sports and Culture of Japan, by a

www.eje.org

Downloaded from Bioscientifica.com at 09/29/2021 03:55:47PM via free access EUROPEAN JOURNAL OF ENDOCRINOLOGY (2003) 149 Adrenomedullin and resistin in adipocytes 237

Research Grant from the HIROMI Medical Research 19 Moore GB, Chapman H, Holder JC, Lister CA, Piercy V, Smith SA Foundation (2001) and by a Research Grant from the et al. Differential regulation of adipocytokine mRNAs by rosiglitazone in db/db mice. Biochemical and Biophysical Research Intelligent Cosmos (2002). Communications 2001 286 735–741. 20 Nagaev I & Smith U. Insulin resistance and are not related to resistin expression in human fat cells or skeletal muscle. References Biochemical and Biophysical Research Communications 2001 285 561–564. 1 Zhang Y, Proenca R, Maffei M, Barone M, Leopold L & 21 Janke J, Engeli S, Gorzelniak K, Luft FC & Sharma AM. Resistin Friedman JM. Positional cloning of the mouse obese gene and gene expression in human adipocytes is not related to insulin its human homologue. Nature 1994 372 425–432. resistance. Obesity Research 2002 10 1–5. 2 Berg AH, Combs TP, Du X, Brownlee M & Scherer M. The adipo- 22 Savage DB, Sewter CP, Klenk ES, Segal DG, Vidal-Puig A, cyte-secreted protein Acrp30 enhances hepatic insulin action. Considine RV et al. Resistin/Fizz3 expression in relation to obesity Nature Medicine 2001 7 947–953. and peroxisome proliferator-activated receptor-gamma action in 3 Maeda K, Okubo K, Shimomura I, Funahashi T, Matsuzawa Y & humans. Diabetes 2001 50 2199–2202. Matsubara K. cDNA cloning and expression of a novel adipose 23 Kim KH, Lee K, Moon YS & Sul HS. A -rich adipose tissue- specific -like factor, apM1 (AdiPose Most abundant Gene specific secretory factor inhibits adipocyte differentiation. Journal transcript 1). Biochemical and Biophysical Research Communications of Biological Chemistry 2001 276 11252–11256. 1996 221 286–289. 24 Kroder G, Bossenmaier B, Kellerer M, Capp E, Stoyanov B, 4 Hotamisligil GS, Shargill NS & Spiegelman BM. Adipose Muhlhofer A et al. Tumor necrosis factor-alpha- and hyper- expression of tumor necrosis factor-alpha: direct role in obesity- glycemia-induced insulin resistance. Evidence for different linked insulin resistance. Science 1993 259 87–91. mechanisms and different effects on insulin signaling. Journal of 5 Kitamura K, Kangawa K, Kawamoto M, Ichiki Y, Nakamura S, Clinical Investigation 1996 97 1471–1477. Matsuo H et al. Adrenomedullin: a novel hypotensive peptide 25 Fasshauer M, Klein J, Neumann S, Eszlinger M & Paschke R. isolated from human pheochromocytoma. Biochemical and Isoproterenol inhibits resistin gene expression through a G(S)- Biophysical Research Communications 1993 192 553–560. protein-coupled pathway in 3T3-L1 adipocytes. FEBS Letters 6 Shimosawa T, Ogihara T, Matsui H, Asano T, Ando K & Fujita T. 2001 500 60–63. Deficiency of adrenomedullin induces insulin resistance by 26 Fasshauer M, Klein J, Neumann S, Eszlinger M & Paschke R. Tumor increasing oxidative stress. Hypertension 2003 41 1080–1085. necrosis factor alpha is a negative regulator of resistin gene 7 Hayashi M, Shimosawa T, Isaka M, Yamada S, Fujita R & Fujita T. expression and secretion in 3T3-L1 adipocytes. Biochemical and Plasma adrenomedullin in diabetes. Lancet 1997 350 Biophysical Research Communications 2001 288 1027–1031. 1449–1450. 27 Shojima N, Sakoda H, Ogihara T, Fujishiro M, Katagiri H, Anai M 8 Martinez A, Elsasser TH, Bhathena SJ, Pio R, Buchanan TA, et al. Humoral regulation of resistin expression in 3T3-L1 and Macri CJ et al. Is adrenomedullin a causal agent in some cases mouse adipose cells. Diabetes 2002 51 1737–1744. of type 2 diabetes? Peptides 1999 20 1471–1478. 28 Green H & Kehinde O. An established preadipose cell line and its 9 Martinez A, Weaver C, Lopez J, Bhathena SJ, Elsasser TH, Miller MJ differentiation in culture. II. Factors affecting the adipose et al. Regulation of insulin secretion and blood glucose metab- olism by adrenomedullin. Endocrinology 1996 137 2626–2632. conversion. Cell 1975 5 19–27. 10 Pio R, Martinez A & Cuttitta F. Cancer and diabetes: two patho- 29 Green H & Kehinde O. Spontaneous heritable changes leading logical conditions in which adrenomedullin may be involved. to increased adipose conversion in 3T3 cells. Cell 1976 7 Peptides 2001 22 1719–1729. 105–113. 11 Takahashi K. Adrenomedullin from a pheochromocytoma to the 30 Takahashi K, Satoh F,Hara E, Murakami O, Kumabe T, Tominaga T eye: implications of the adrenomedullin research for endocrin- et al. Production and secretion of adrenomedullin by cultured ology in the 21st century. Tohoku Journal of Experimental Medicine choroid plexus carcinoma cells. Journal of Neurochemistry 1997 2001 193 79–114. 68 726–731. 12 Sugo S, Minamino N, Kangawa K, Miyamoto K, Kitamura K, 31 Sakata J, Shimokubo T, Kitamura K, Nakamura S, Kangawa K, Sakata J et al. Endothelial cells actively synthesize and secrete adre- Matsuo H et al. Molecular cloning and biological activities of rat nomedullin. Biochemical and Biophysical Research Communications adrenomedullin, a hypotensive peptide. Biochemical and Biophysical 1994 201 1160–1166. Research Communications 1993 195 921–927. 13 Sugo S, Minamino N, Shoji H, Kangawa K, Kitamura K, Eto T et al. 32 Totsune K, Mackenzie HS, Totsune H, Troy JL, Lytton J & Production and secretion of adrenomedullin from vascular Brenner BM. Upregulation of atrial natriuretic peptide gene smooth muscle cells: augmented production by tumor necrosis expression in remnant of with reduced renal mass. factor-alpha. Biochemical and Biophysical Research Communications Journal of the American Society of Nephrology 1998 9 1613–1619. 1994 203 719–726. 33 Satoh F, Takahashi K, Murakami O, Totsune K, Sone M, Ohneda M 14 Isumi Y, Minamino N, Katafuchi T, Yoshioka M, Tsuji T, et al. Adrenomedullin in human brain, adrenal glands and tumor Kangawa K et al. Adrenomedullin production in fibroblasts: its tissues of pheochromocytoma, ganglioneuroblastoma and neuro- possible function as a growth regulator of Swiss 3T3 cells. blastoma. Journal of Clinical Endocrinology and Metabolism 1995 Endocrinology 1998 139 2552–2563. 80 1750–1752. 15 Satoh F, Takahashi K, Murakami O, Totsune K, Sone M, Ohneda M 34 Nishimatsu H, Suzuki E, Nagata D, Moriyama N, Satonaka H, et al. Immunocytochemical localization of adrenomedullin-like Walsh K et al. Adrenomedullin induces endothelium-dependent immunoreactivity in the human and the adrenal vasorelaxation via the phosphatidylinositol 3-kinase/Akt-depen- gland. Neuroscience Letters 1996 203 207–210. dent pathway in rat aorta. Circulation Research 2001 89 63–70. 16 Steppan CM, Bailey ST, Bhat S, Brown EJ, Banerjee RR, Wright CM 35 Way JM, Gorgun CZ, Tong Q, Uysal KT, Brown KK, et al. The hormone resistin links obesity to diabetes. Nature 2001 Harrington WWet al. Adipose tissue resistin expression is severely 409 307–312. suppressed in obesity and stimulated by peroxisome proliferator- 17 Banerjee RR & Lazar MA. Dimerization of resistin and resistin-like activated receptor gamma . Journal of Biological Chemistry molecules is determined by a single cysteine. Journal of Biological 2001 276 25651–25653. Chemistry 2001 276 25970–25973. 36 Feinstein R, Kanety H, Papa MZ, Lunenfeld B & Karasik A. Tumor 18 Haugen F, Jorgensen A, Drevon CA & Trayhurn P. Inhibition by necrosis factor-alpha suppresses insulin-induced tyrosine phos- insulin of resistin gene expression in 3T3-L1 adipocytes. FEBS phorylation of insulin receptor and its substrates. Journal of Letters 2001 507 105–108. Biological Chemistry 1993 268 26055–26058.

www.eje.org

Downloaded from Bioscientifica.com at 09/29/2021 03:55:47PM via free access 238 Y Li and others EUROPEAN JOURNAL OF ENDOCRINOLOGY (2003) 149

37 Hotamisligil GS, Peraldi P, Budavari A, Ellis R, White MF & 45 McLatchie LM, Fraser NJ, Main MJ, Wise A, Brown J, Thompson N Spiegelman BM. IRS-1-mediated inhibition of insulin receptor et al. RAMPs regulate the transport and ligand specificity of the tyrosine kinase activity in TNF-alpha- and obesity-induced insulin calcitonin-receptor-like receptor. Nature 1998 393 333–339. resistance. Science 1996 271 665–668. 46 Nagae T, Mukoyama M, Sugawara A, Mori K, Yahata K, 38 Uysal KT, Wiesbrock SM, Marino MW & Hotamisligil GS. Protec- Kasahara M et al. Rat receptor-activity-modifying proteins tion from obesity-induced insulin resistance in mice lacking (RAMPs) for adrenomedullin/CGRP receptor: cloning and upre- TNF-alpha function. Nature 1997 389 610–614. gulation in obstructive nephropathy. Biochemical and Biophysical 39 Bedard S, Marcotte B & Marette A. modulate glucose Research Communications 2000 270 89–93. transport in skeletal muscle by inducing the expression of induci- 47 Keay S & Grossberg SE. Interferon inhibits the conversion of 3T3- ble nitric oxide synthase. Biochemical Journal 1997 325 L1 mouse fibroblasts into adipocytes. PNAS 1980 77 4099–4103. 487–493. 48 Doerrler W, Feingold KR & Grunfeld C. Cytokines induce catabolic 40 Stephens JM, Lee J & Pilch PF. Tumor necrosis factor-alpha- effects in cultured adipocytes by multiple mechanisms. Cytokines induced insulin resistance in 3T3-L1 adipocytes is accompanied 1994 6 478–484. by a loss of insulin receptor substrate-1 and GLUT4 expression 49 Gregoire F, De Broux N, Hauser N, Heremans H, Van Damme J & without a loss of insulin receptor-mediated signal transduction. Remacle C. Interferon-gamma and interleukin-1 beta inhibit adi- Journal of Biological Chemistry 1997 272 971–976. poconversion in cultured rodent preadipocytes. Journal of Cellular 41 Stephens JM & Pekala PH. Transcriptional repression of the Physiology 1992 151 300–309. GLUT4 and C/EBP in 3T3-L1 adipocytes by tumor necrosis 50 Grossberg SE & Keay S. The effects of interferon on 3T3-L1 cell factor-alpha. Journal of Biological Chemistry 1991 266 differentiation. Annals of the New York Academy of Sciences 1980 21839–21845. 350 294–300. 42 Takahashi K, Nakayama M, Totsune K, Murakami O, Sone M, 51 Waite KJ, Floyd ZE, Arbour-Reily P & Stephens JM. Interferon- Kitamuro T et al. Increased secretion of adrenomedullin from gamma-induced regulation of peroxisome proliferator-activated cultured human astrocytes by cytokines. Journal of Neurochemistry receptor gamma and STATs in adipocytes. Journal of Biological 2000 74 99–103. Chemistry 2001 276 7062–7068. 43 Takahashi K, Satoh F, Hara E, Sone M, Murakami O, Kayama T 52 Memon RA, Feingold KR, Moser AH, Fuller J & Grunfeld C. Regu- et al. Production and secretion of adrenomedullin from glial cell lation of fatty acid transport protein and fatty acid translocase tumors and its effects on cAMP production. Peptides 1997 18 mRNA levels by endotoxin and cytokines. American Journal of 1117–1124. Physiology 1998 274 E210–E217. 44 Udono T, Takahashi K, Nakayama M, Murakami O, Durlu YK, Tamai M et al. Adrenomedullin in cultured human retinal pigment epithelial cells. Investigative Ophthalmology and Visual Received 7 February 2003 Science 2000 41 1962–1970. Accepted 3 June 2003

www.eje.org

Downloaded from Bioscientifica.com at 09/29/2021 03:55:47PM via free access